Whisper-AT combines speech recognition and audio tagging for improved performance.
― 5 min read
James Glass
A new model improves understanding of speech and sounds simultaneously.
― 6 min read
A new method enhances testing robustness of language models by prioritizing novelty.
― 7 min read
ThReaD improves LLMs' performance on complex tasks through dynamic thread management.
― 5 min read
Self-MoE creates specialized experts for improved language model performance.
― 6 min read
A look at the Codec-SUPERB challenge results and codec performance metrics.
― 5 min read
Machines learn to locate objects in images using innovative techniques.
― 5 min read